{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreiceiuekjk2n6gaelhkvwlgghbbru6cp6npefwmq2yx5gxar4axjn4",
"uri": "at://did:plc:ivbknywyskln22er3nkssdhl/app.bsky.feed.post/3mo622w2f5rf2"
},
"path": "/t/named-impl-with-implementation-selection-variant/24374#post_9",
"publishedAt": "2026-06-13T09:15:22.000Z",
"site": "https://internals.rust-lang.org",
"tags": [
"the hashtable problem",
"facet"
],
"textContent": "viruscamp:\n\n> 2. […] at least when a default implementation exists.\n>\n\n\nThis breaks adding or broadening a trait `impl` being SemVer-safe.\n\nviruscamp:\n\n> 1. By forbidding named implementations for specific traits, the hashtable problem is cleanly bypassed.\n>\n\n\nThe \"hashtable problem\" is a stand-in for the general category of traits with consistency requirements. This includes third-party definitions like the (de)serialisation traits in Serde or the reflection traits in facet (in combination with type keying).\n\nIt's not quite a soundness problem, but the logical consistency aspect here is important enough that, in my opinion, no proposal that special-cases specific traits can solve it ergonomically.",
"title": "Named impl with Implementation Selection Variant"
}